AI is gaining ground
As far as hot topics go, artificial intelligence is one of the biggest these days. Everyone’s talking about it. The tech sector is influencing pretty much every single industry, and the building and construction fields are starting to catch on.
It’s widely known (and lamented) that the various building industries have a bit of a coordination problem when it comes to technology. And isn’t that a bit ironic? Architects, planners, and developers are trained to coordinate many moving parts, after all. We’re getting on board with new technologies. Slowly, but surely, we’re picking up on things like AI over time.
Speaking of time, we’ve had BIM for ages now. BIM has helped architects, engineers, developers, contractors, and building consultants collaborate better and start speaking the same language. BIM has improved both design processes and designs themselves.
Especially in the building industries, it can be tempting to turn to the latest thing. But that shouldn’t come at the expense of comprehensive, helpful technologies like BIM. Using models to our utmost advantage and leveraging their power in tandem with rising innovations will be critical as we advance into a world where artificial intelligence is the norm.
While it can be said that thanks to BIM and other software, we’re designing and building more efficiently than ever, we’re not designing and building as efficiently as possible. That’s where AI really comes in.
Where does implementing AI begin?
BIM and building simulation software yield so much data that most people don’t know what to do with it. Architecture and consulting firms around the world have created entire divisions dedicated to figuring out how to collect, organize, store, and share building data. Using data collected from simulations, models, and even physical elements like sensors inside completed constructions, we can actually transform the design process, or at the very least, innovate with each new building project.
Industry expert Nicholas Klokhol, a director at Geniebolt recently discussed the buzz around BIM, big data, and AI in the building and construction sectors. He outlines some problems we have with data and strategies for working through them. “We generate a large amount of the data but once a project is done, we throw away 95% of all the data that we generated instead of analysing and automating to optimise not only our own business, but also the projects we are working on,” he states.
This is a very commonly held view; that architects, engineers, and other building professionals don’t use data to their advantage. The tendency to move on to the next project without thinking how all the previously collected data can be used to enhance it.
Understanding how AI works
Once we compile and organize data, figuring out how to use it properly and really take advantage of computation is the next step. In AI, there are a number of different kinds of algorithms that use data to generate solutions for complicated problems. One type that is particularly suited for building design challenges is the evolutionary algorithm. As its name would suggest, an evolutionary algorithm imitates biological evolution, where the components evolve to present an optimal solutions or options.
In a recently published article, “Artificial Intelligence Aided Architectural Design,” Jan Cudzik and Kacper Radziszewski discuss different methods of implementing intelligent-based algorithms to improve design practices. They indicate that evolutionary algorithms “are becoming the matter of interest for artists, designers and architects,” adding that their applications in architectural design have been long-studied. They note that evolutionary algorithms can help designers solve problems by optimizing things like structural topology, cross-sections, and materials selection.
Perhaps the most exciting opportunity with intelligent evolutionary algorithms lies in the early design phases. METABUILD’s algorithm, for example, assesses models and other data related to energy performance, lighting and thermal comfort, and air quality to generate design options that meet or exceed sustainability goals. Better yet, it assesses cost performance factors to limit life-cycle expenses and maintain cost-effectiveness.
Designs that really take sustainability, comfort, and cost into serious consideration will produce better buildings. And we can generate those designs with the help of AI.